Add guidance for adding new metrics

[BrynCooke]

Until now there has been little guidance on adding new metrics to the router. This PR expands the dev doc to include this.

---

Checklist

Complete the checklist (and note appropriate exceptions) before the PR is marked ready-for-review.

• Changes are compatible¹
• Documentation² completed
• Performance impact assessed and acceptable
• Tests added and passing³
  • Unit Tests
  • Integration Tests
  • Manual Tests

Exceptions

Note any exceptions here

Notes

Footnotes

1. It may be appropriate to bring upcoming changes to the attention of other (impacted) groups. Please endeavour to do this before seeking PR approval. The mechanism for doing this will vary considerably, so use your judgement as to how and when to do this. ↩

2. Configuration is an important part of many changes. Where applicable please try to document configuration examples. ↩

3. Tick whichever testing boxes are applicable. If you are adding Manual Tests, please document the manual testing (extensively) in the Exceptions. ↩

[github-actions]

@BrynCooke, please consider creating a changeset entry in /.changesets/. These instructions describe the process and tooling.

[router-perf]

CI performance tests

• step - Basic stress test that steps up the number of users over time
• events_big_cap_high_rate_callback - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity using callback mode
• large-request - Stress test with a 1 MB request payload
• events - Stress test for events with a lot of users and deduplication ENABLED
• xxlarge-request - Stress test with 100 MB request payload
• events_without_dedup - Stress test for events with a lot of users and deduplication DISABLED
• xlarge-request - Stress test with 10 MB request payload
• step-jemalloc-tuning - Clone of the basic stress test for jemalloc tuning
• events_callback - Stress test for events with a lot of users and deduplication ENABLED in callback mode
• no-graphos - Basic stress test, no GraphOS.
• reload - Reload test over a long period of time at a constant rate of users
• events_big_cap_high_rate - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity
• events_without_dedup_callback - Stress test for events with a lot of users and deduplication DISABLED using callback mode
• const - Basic stress test that runs with a constant number of users

[Geal]

a lot of what is encoded in this document is unclear to me, I think we should discuss it a bit more

[Geal]

Suggested change

	* Static - Used by us to monitor feature usage.
	* Static - Used by Router developers to monitor feature usage.

[Geal]

let's assume router users will end up looking at the dev docs

[Geal]

that is not clear to me. What do we mean by "users using static metrics directly?" Is it when they would add that in their custom plugin? (which would not increase cardinality for all users) Or asking us to add a new metric to the router?

[Geal]

a lot of static metrics actually monitor standard router operations and are not for us to collect data, but for users to observe the router.
If we want this to be the defining point, let's maybe not call them static VS dynamic metrics, but internal VS monitoring or user metrics, something like that?
I'd prefer we keep the distinction between static metrics as defined directly with tracing, and dynamic metrics as the ones defined by custom instruments, that can be activated with runtime conditions, and have another clear separation beween the metrics used for internal reporting (as with the apollo.router.operations and apollo.router.config prefixes) and the user facing ones.

[Geal]

code example of what is a static metric, to be sure which is which between static and dynamic?

[Geal]

some of the apollo.router.operations metrics are actually monitored by users. What is the strategy here? Do we keep them available for users?